I’ve been reading some interesting opinion on OnLive from various websites – for example Richard Leadbetter at Eurogamer (here). His main criticisms of the announcement are the amount of artefacts you would see running 720p video at 60fps across the wire (the guy works for Digital Foundry), the amount of hardware needed in the back-end and latency. I’m not saying that I’ve been converted and I see OnLive taking over the gaming world by the end of the year, but I do think you need to think out the box here and stop thinking about this in relation to the way things work now, with powerful consoles etc..

Let’s take his points:

• Video compression – it’s highly likely there will be artefacts involved here – however he also shows that 30fps SD video looks pretty damn good. So things are already possible on a certain level. Things can only get better from there – think about how much video compression and streaming technologies have progressed in the last 5-10 years. Who’s to say there won’t be just as many strides in the next 5-10 years? He said his colleagues reactions were often laughing at the thought of what OnLive are claiming, but why can’t a company come along that has made some real innovations? Just because you are an expert in the field at this moment, doesn’t mean someone won’t come along at some point and re-write the rulebook.
• Latency – he admits that with an IPTV-style model this could give you enough bandwidth in theory but then talks about how much ISP’s complained about the bandwidth used by BBC Iplayer and how they reacted to that, but again, where have we come from? Imagine the thought of streaming TV over your net connection when we were using 14.4K modems! Things have progressed and adapted and I see no reason why things won’t change and adapt in the future. If you just look at business models of ISPs now and how things are currently set up, it does seem like a service such as OnLive could struggle with real world conditions but things evolve. I see no reason why OnLive won’t be another reason for ISPs to evolve further.
• Hardware – this is probably the biggest point he raised. If thousands of people want to start playing GTAV on the day of release how are they going to cope with that? Think of the thousands of servers needed, how powerful they’d need to be and then think of the costs!…but again, think outside the box here. People are thinking about just adapting the same game that runs on a console or PC to run on a server, but as this technology grows code will change to be more efficient on this kind of platform. New hardware technology will be produced for this exact sort of use once demand grows.

An interesting article at PALGN puts a more positive slant on things here. They quote an article on game response times, i.e. input lag – with Virtua Tennis 3 having a 67ms response time and GTAIV 167ms. The report concluded that anything beyond 167ms felt sluggish. So if OnLive are building datacenters which are up to 1000 miles away, it’s going to create latencies of around 30ms. Video compression technology has been pretty slow (500ms+) but the technology OnLive seem to be trumpeting is the fact that they can do the same thing in 1ms – and why not? They have been working on the technology for 7 years – maybe this is where their breakthrough has been. So add another 5ms or so on to the total figure for decompression at the client end and you could be talking sub 40ms. This is still well within the game response times people experience on a console, so in theory it all seems feasible.

Maybe OnLive is just sowing the seeds. Maybe it will fail as it’s still a bit too early for their business model to work, especially where hardware costs are involved, but I don’t think this is a flash in the pan. as I keep saying, the key is to think outside the box – many technologies we take for granted now were laughed at when they first came on the scene. ‘It’ll never work’ people say. Well, sometimes things don’t quite live up to the claims but generally as technologies mature they do in the end. The whole concept of OnLive is a rule-breaker. People comment about how Sony or Microsoft will think about this, and how they won’t want to stop their lucrative console/game tie-ins…but that’s thinking about the whole thing with today’s rules. Rip up the rule book and things have to change.

PALGN sum it up nicely when they say “OnLive still has a lot it needs to show, but it does look like it has the ability to live up to the hype. Its possible impact on gaming over the next 5 years can’t be understated.”